Starting Tuesday 1st March 2016:
- 18:00-20:00 Tuesdays in LB04
- 18:00-20:00 Thursdays in LB04
Playing with data is an important part of Statistics modules. Help is given on this webpage for using R.
Some lecturenotes for a similar course (ST3010 run for undergrads in Michaelmas term) are available. ST7005 follows the same content as ST3010 in a shorter time frame, hence in ST7005 there will be less emphasis on some of the mathematical content of the lecturenotes and more emphasis on practical aspects.
- Exam (100%)
There are a lot of books on time series, forecasting in the Library that are relevant to the course. Below the first book is given as an example reference. It is available in the library and most of the time series used in the labs are explained in that book.
01/03/2016: Introduction to Time series, Visualisation of time series (lecturenotes chapters 1&2)
Prediction is very difficult, especially about the future:
- Example with Earthquake
- Example of using forecasting techniques: Detecting influenza epidemics using search engine query data, Nature 2009
- Example of misusing forecasting techniques: The pending demise of facebook (and Princeton) or Facebook 's Future
03/03/2016: Holt Winters Algorithms: SES and DES (chapters 4,5 and 6)
08/03/2016: Seasonal Holt Winters Algorithms
- Excel SES and DES algorithms for Dowjones time series
- Excel SHW+ algorithm for beer time series.
- The Holt-Winters Approach to Exponential Smoothing: 50 Years Old and Going Strong, Paul Goodwin, Foresight Fall 2010
10/03/2016: Linear Regression and Auto Regressive models (chapter 8 and 9)
- 15/03/2016: ARIMA(p,d,q) (chapters 9, 10, 11)
17/03/2016: Bank holiday
- 22/03/2016: Seasonal ARIMA (LB04)
24/03/2016: Lab on HoltWinters Algorithms in East End PC labs (EE.PC)
- 29/03/2016: Backshift operator, stationarity in variance (LB04)
31/03/2016: Exercises/Exam revisions (LB04)
- 05/04/2016: Conclusions (LB04)
- 07/04/2016: Lab on ARIMA models in East End PC labs (EE.PC)